Virulence Reduction in Yersinia pestis by Combining Delayed Attenuation with Plasmid Curing

TL;DR

Researchers developed a safer Yersinia pestis strain by combining delayed attenuation with plasmid removal, showing strong protection in mice and guinea pigs.

Contribution

A novel approach combining delayed attenuation and plasmid curing in Yersinia pestis to create a promising vaccine candidate.

Findings

The 231Δcrp strain was nonlethal in mice at high doses and increased LD50 in guinea pigs by 107-fold.

The 231PBAD-crp(pPst¯) strain showed no mortality in mice and guinea pigs and induced strong protective immunity.

The study highlights the balance between safety and effectiveness in developing plague vaccines.

Abstract

Yersinia pestis caused the three plague pandemics that claimed more than two hundred million human lives. There is still no vaccine that meets all WHO requirements, and many researchers continue to develop plague vaccines using various technological platforms. For example, researchers led by Roy Curtiss 3rd have developed a new approach to achieve controlled, delayed attenuation of bacterial pathogens. Mutants generated using this method were superior in protecting Y. pestis-infected mice immunized with strains generated using traditional gene knockout. However, further studies are needed to determine the safety and efficacy of these delayed-attenuated strains in other mammalian species in order to extrapolate on humans the data obtained in accordance with the FDA Animal Rule. Three Y. pestis strains, a Δcrp mutant, a mutant with arabinose-dependent regulated crp expression (araC PBAD crp) or an araC PBAD crp mutant cured of plasmid pPst were derived from virulent wild-type strain 231. To evaluate the safety, outbred mice or guinea pigs were immunized subcutaneously with serial tenfold dilutions of mutated strains. For vaccine studies, immunized animals were subcutaneously challenged with 200 LD100 (lethal dose in all exposed subjects) of the wild-type Y. pestis strain. The challenge caused the death of 100% of naïve animals in controls. The Y. pestis strain 231Δcrp was nonlethal in mice at a dose of 107 CFs. The LD50 of the 231Δcrp strain in guinea pigs increased by at least 107-fold compared to that of the wild-type strain. The LD50s of the 231PBAD-crp mutant in mice and guinea pigs were approximately 104-fold and 107-fold higher than those of Y. pestis 231, respectively. The 231PBAD-crp(pPst¯) strain did not cause death in mice (LD50 > 107 CFU) and guinea pigs (LD50 > 109 CFU) when administered subcutaneously and was capable of inducing intense protective immunity in both species of laboratory animals. Our research has shown once again the necessity of balance between safety and effectiveness demonstrating the feasibility of further investigation of crp mutants as promising candidate plague vaccines.

Figures

Full text

1. Introduction

Plague is a disease of wild rodents, which only accidentally crosses over to humans [1]. However, these accidents were the cause of at least three known pandemics and cost humanity more than 200 million lives [2,3]. The World Health Organization (WHO), a key player in pandemic preparedness and response [4], has released a list of prioritized pathogens posing the greatest emerging epidemic threats. WHO draws the attention of the global community to the need to accelerate research in the development and clinical trials of vaccines against these priority pathogens [5,6]. This list contains five bacterial pathogens, including Yersinia pestis, the causative agent of the plague. Although the success in the fight against the third plague pandemic was largely due to the application of the first generation of plague vaccines (e.g., a killed plague vaccine developed in 1897 by Waldemar Mordecai Haffkine (1860–1930)) [7,8] there is still no generally accepted plague vaccine that meets all WHO requirements [9]. A number of researchers continue to develop plague vaccines using various technological platforms [10,11]. Advances in molecular bacteriology have made it possible to develop a wide range of plague vaccine candidates that appear to be superior to the first-generation vaccines in vitro and in animal comparative studies [12]. Recently, Roy Curtiss 3rd’s team proposed an original methodology for controlled delayed attenuation of bacterial pathogens [13,14,15]. This technology has been applied to Y. pestis strain KIM5+ (subsp. pestis bv. Medievalis) in a mouse model and has shown superiority over traditional knockout mutagenesis [16]. However, the safety and efficacy of the strains with delayed attenuation in humans is not yet known.

Currently, ethical considerations prohibit the use of standard research methodologies like randomized controlled trials or cohort studies for evaluating plague vaccines. This is because failure to vaccinate the control group would expose its members to unacceptable risk. Therefore, the authorization of plague vaccines for use in humans should be conducted utilizing the FDA’s Animal Rule [17]. This rule facilitates the licensing of products for which traditional potency testing methodologies are deemed unethical. Specifically, the Animal Rule is applied in situations where an infective agent exhibits extremely high pathogenicity and the rarity of the infection it causes precludes the feasibility of conducting field studies (21 CFR 601.91 Subpart H, “Approval of Biological Products When Human Efficacy Studies Are Not Ethical or Feasible”) [18]. One of the four essential requirements for the application of the FDA Animal Rule is that the vaccine’s effects (survival or prevention of serious disease) must be observed in more than one animal species, which is expected to be more likely to reflect human responses.

The aim of this study was to verify the vaccine properties of the P_BAD_-crp mutant generated on the basis of a phylogenetically different parent strain and to evaluate its properties not only in mice but also in guinea pigs.

2. Materials and Methods

2.1. Bacterial Strains, Plasmids, and Culture Conditions

All bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli DH5α was used for plasmid propagation. Y. pestis cells were grown routinely at 28 °C in BHI, HiMedia, Mumbai, India) or on BHI agar at pH 7.2. E. coli cells were grown routinely at 37 °C in Luria–Bertani (LB) broth (HiMedia, Mumbai, India) or on agar plates at pH 7.2. Ampicillin (100 µg/mL, Amp) (neoFroxx, Einhausen, Germany), chloramphenicol (20 µg/mL, Cm) (neoFroxx, Einhausen, Germany), polymyxin B (50 μg/mL, Pol) (neoFroxx, Einhausen, Germany), L-arabinose (200 µg/mL, Ara) (HiMedia, Dahej, India) were supplemented as appropriate. BHI agar containing 10% sucrose was used for sacB-based counterselection in the allelic exchange experiments. MacConkey agar (HiMedia, Mumbai, India) with 0.5% maltose (HiMedia, Dahej, India) was used to indicate sugar fermentation and crp expression.

2.2. Animals

Six-week-old outbred mice (Lab of Biomodels breeding unit, SCRAMB, Moscow, Russia) and four-week-old guinea pigs («Andreevka» branch of the Scientific Center of Biomedical Technologies of the Federal Medical and Biological Agency (Andreevka, Moscow, Russia) both male and female were used in all studies. All animal experiments were approved by the State Research Center for Applied Microbiology and Biotechnology (Obolensk, Moscow, Russia) Bioethics Committee (Permit No: VP-2025/1 on 30 April 2025) and were performed in compliance with the NIH Animal Welfare Insurance #A5476-01 issued on 2 July 2007, and the European Union guidelines and regulations on handling, care and protection of Laboratory Animals (https://environment.ec.europa.eu/topics/chemicals/animals-science_en (accessed on 18 November 2025).

2.3. Construction of the Plasmids

The primers used in this work are listed in Table 2.

A primer set of Crp-NdeI/Crp-XhoI was used for amplifying the gene sequence of crp. The resulting PCR product was digested with NdeI and XhoI and ligated with plasmid pET-24b(+) digested with the same restriction enzymes to construct plasmid pET24-crp. E. coli BL21(DE3) was transformed with the plasmid pET24-crp.

Primer sets of Crp-HindIII/Crp-SphI and Crp-XbaI/Crp-SalI were used for amplifying URcrp (upstream region gene sequence of crp, from −664 to −110 bp) and the gene sequence of crp with the Shine-Dalgarno sequence, from −36 to +633 bp, respectively. A primer set of Pbad-SphI/Pbad-XbaI was used for amplifying the araC P_BAD_ promoter from strain E. coli DH5α. The resulting PCR products were digested with appropriate restriction enzymes and ligated with pUC57 digested with HindIII and SalI to construct the plasmid pUC57-URcrp-araC Pbad-crp.

A primer set of Lt0-SphIF/Lt0-SphIR was used for amplifying the transcription terminator t0 from bacteriophage λ (Lt0TT). The resulting PCR product was digested with SphI and ligated with the plasmid pUC57-URcrp-araC Pbad-crp digested with the same restriction enzyme to construct the plasmid pUC57-URcrp-Lt0TT-araC Pbad-crp. The plasmid pUC57-URcrp-Lt0TT-araC Pbad-crp was digested with SmaI and ligated with the suicide vector pCVD442 digested with the same restriction enzyme to construct the plasmid pCVD442-URcrp-Lt0TT-araC Pbad-crp.

All the plasmid constructions were verified through sequencing.

2.4. Mutagenesis

The crp deletion was constructed in the Y. pestis EV strain by λRed-mediated mutagenesis [21] and confirmed by PCR (Table 2). The Y. pestis EVΔcrp::cat DNA fragment containing the respective deletion was then subcloned into the pCVD442 plasmid [22]. The pCVD442-Δcrp::cat plasmid was transferred from an E. coli S17-1 λpir donor strain into a recipient wild type Y. pestis strain 231 by conjugation. Elimination of the suicide vector and selection of Y. pestis clones were performed in the presence of 5% sucrose and chloramphenicol [22]. The resistance cassette was eliminated to produce FRT scar mutants by introducing pCP20 [21].

The procedure for constructing the Y. pestis 231P_BAD_-crp mutant was performed using the standard method [22]. In brief, the suicide plasmid pCVD442-URcrp-Lt0TT- araC P_BAD_-crp was conjugationally transferred from E. coli S17-1 λpir to Y. pestis 231. A single-crossover insertion strain was isolated on BHI agar plates containing ampicillin. Loss of the suicide vector after the second recombination between the homologous regions was selected using the sacB-based sucrose sensitivity counter-selection system. The colonies were screened for Amp^S^ and verified by PCR using the primers Crp-KF/Crp-KR.

The isogenic pPst-plasmid-cured strain of Y. pestis 231P_BAD_-crp(pPst¯) was generated as described earlier. It was cloned from the 231P_BAD_-crp strain after 10 serial passages caused in BHI broth media at +5 °C [24]. Clones lacking the 9.5-kb pPst plasmid were detected by PCR and confirmed by a fibrinolysis assay [25].

The presence of all Y. pestis resident plasmids [26,27,28,29] was confirmed via PCR amplification.

2.5. Preparation of Crp Antiserum

Full-length His-tagged Crp was expressed from E. coli BL21(DE3)/pET24-crp. Briefly, recombinant E. coli culture was grown to log phase in LB broth at 37 °C with agitation. Cells were then induced with 1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) for 4 h. After sonication and centrifugation, protein was solubilized in the binding buffer containing 6 M urea and was purified in an affinity chromatography using Ni^2+^-Sepharose chromatography under denaturing conditions.

Mouse anti-CRP serum was collected after two subcutaneous immunizations with 20 μg per animal after being mixed with the Imject Alum (1:1).

2.6. Western Blotting

The whole-cell lysates were separated by sodium dodecyl sulfate-polyacrylamide (10%) gel electrophoresis and then electrophoretically transferred to a nitrocellulose membrane using equipment and protocols from Bio-Rad Laboratories (Hercules, CA, USA). The membrane was blocked overnight at 4 °C with 5% (w/v) nonfat dry milk in phosphate-buffered saline containing 0.2% Tween 20 (PBS-T), washed three times for 10 min with PBS-T, probed at room temperature for 1.5 h with a mice polyclonal against His-tagged Crp from E. coli BL21(DE3)/pET24-crp diluted in PBS-T, and then washed three times with PBS. Immunoreactive proteins were detected using a horseradish peroxidase-conjugated anti-mouse IgG-HRP (1:5000, Sigma, Saint-Louis, MO, USA). Proteins were visualized with DAB substrate kit (Abcam, Cambridge, UK) and photographed using an Amersham ImageQuant 800 (Cytiva, Wilmington, NC, USA) gel imaging system.

2.7. Animal Challenges

Y. pestis strains were grown at 28 °C for 48 h on BHI plus L-arabinose (200 µg/mL), diluted to an appropriate concentration in PBS. The challenge doses were determined by serial dilutions in PBS and plating on BHI agar.

To demonstrate the loss of virulence, mice (n = 6/per dose) and guinea pigs (n = 6/per group) were challenged subcutaneously (s.c.) with the 231Δcrp, 231P_BAD_-crp or 231P_BAD_-crp(pPst¯) strains. Each strain was inoculated at different 10-fold doses at 0.2 mL aliquots (10^2^ CFU–10^7^ CFU per mouse, 10^2^–10^9^ CFU per guinea pig). At the same time, 4 additional groups of 6 mice or guinea pigs were used to control the virulence of the wild type Y. pestis 231. Then the animals were monitored for 28 days to assess the survival rate.

For vaccine studies, groups of 6 outbred mice or 6 guinea pigs were vaccinated via the subcutaneous (s.c. route with serial dilutions of the Y. pestis 231 strain P_BAD_-crp(pPst¯) or only with PBS buffer as a placebo. Blood was collected on day 28 postimmunization for antibody measurement. On day 30, animals were injected s.c. with 200 LD_100_ (400 CFU for mice, 6 × 10^3^ CFU for guinea pigs) of the wild-type Y. pestis strain 231. All animals were observed over a 30-day period.

2.8. ELISA

ELISA was used to assay serum IgG antibodies against recombinant LcrV [30] and F1 [31] proteins and Y. pestis 231 whole-cell lysate (YPL) [32]. Polystyrene 96-well flat-bottom microtiter plates (Dynatech Laboratories Inc., Chantilly, VA, USA) were coated with 1 μg/well recombinant protein or 0.5 μg/well YPL in 0.3 N sodium carbonate buffer [pH 9.6]. Mouse and guinea pig sera samples were serially diluted from 1:250 to 1:512,000. The endpoint dilution titer was calculated as the serum dilution resulting in an absorbance reading of 0.2 units above background. Goat anti-mouse IgG-HRP (1:5000, Sigma, Saint-Louis, MO, USA) and goat anti-guinea pig IgG-HRP (1:5000, Sigma, Saint-Louis, MO, USA) were used as the detection antibodies. The reactions were developed with TMB (3,3′,5,5′-Tetramethylbenzidine) and stopped with 2 M H_2_SO_4_. The absorbance at 450 nm was measured. Background values were obtained from serum samples collected from the animals injected with the PBS alone.

2.9. In Vivo Cytokine Analysis

Outbred mice (n = 3) and guinea pigs (n = 3) were inoculated s.c. with 10^4^ CFU of Y. pestis 231(P_BAD_-crp)/pPst¯. A group of uninfected mice served as the negative control (0 day). Samples were taken from mice inoculated s.c. on days 2, 4, and 6 post-inoculation. The sera were filtered through 0.22-μm syringe filters (PTFE-0.22-13, JINTENG, Tianjin, China) and cultured on BHI plates to confirm their sterility. The levels of cytokines were measured using an ELISA with Guinea pig Interferon-γ ELISA Kit (#EGP0017), Guinea pig TNF-α ELISA Kit (#EGP0049), Mouse IFN-γ ELISA Kit (#EM0093), and Mouse TNF-α ELISA Kit (#EM0183), all according to the manufacturer’s instructions (Fine Biotech Co., Ltd., Wuhan, China).

2.10. Statistics

The LD_50_ and 95% confidence intervals of the isogenic Y. pestis strains for mice and guinea pigs were calculated using the Kärber method [22]. t-test of unpaired samples, ANOVA and Log-rank (Mantel–Cox) test were used. A p-value below 0.05 was considered significant.

3. Results

3.1. Construction of the Deletion-Insertion Mutation to Achieve Regulated Delayed Attenuation

The Δcrp mutant of Y. pestis strain 231 was generated by replacing the crp with the chloramphenicol resistance gene cat and curing the resistance cassette to produce a FRT scar (Figure 1A). The P_BAD_ crp mutant of Y. pestis strain 231 was constructed by replacing 74 nucleotides containing the crp promoter region from −110 to −37 bp with the transcription terminator t0 from bacteriophage λ (Lt0TT) and the araC P_BAD_ promoter from E. coli strain DH5α (Figure 1B). The deletion of the crp from wild type 231 and the promoter replacement with P_BAD_-crp were confirmed by PCR analysis using specific primers (Table 2) as well as by DNA sequencing of the flanking regions of the crp on the chromosome. The above-mentioned genetic manipulations resulted in creation of the isogenic Δcrp and P_BAD_-crp mutants of Y. pestis strain 231.

The growth of the P_BAD_-crp mutant strain in the presence of arabinose leads to transcription of the crp, but gene expression stopped in the absence of arabinose. The presence of the Crp protein in Y. pestis mutant strains was tested by Western blot using mouse antibodies raised against the His-tagged recombinant protein. Crp was not detected in the either 231Δcrp or 231 P_BAD_-crp mutants grown in the absence of arabinose (Figure 2). After arabinose addition, the P_BAD_-crp strain synthesized approximately the same amount of Crp as the parent wild-type Y. pestis strain 231.

The Δcrp and the P_BAD_-crp strains without arabinose grew more slowly than the parent Y. pestis strain 231 in BHI medium.When 0.2% arabinose was added to the nutrient medium, P_BAD_-crp grew at the same rate as the wild type strain.

Crp increased expression of malT, the transcriptional activator of maltose metabolism [33]. The test of maltose fermentation can readily display the Crp activity. We analyzed the growth of wild-type 231, 231Δcrp, P_BAD_-crp on MacConkey maltose agar without and with 0.2% arabinose (Figure 3). The parent wild-type strain of Y. pestis was able to ferment maltose to acid, which was accompanied by a decrease in pH, and as a result, the colonies changed color on MacConkey medium containing a red indicator with a neutral pH. The Y. pestis 231Δcrp mutant did not grow with maltose as the sole carbon source. The Y. pestis 231P_BAD_-crp strain could ferment maltose only when grown in the presence of arabinose, but not in the absence of it.

3.2. Attenuation of Mutant Strains in s.c. Challenged Mice and Guinea Pigs

To investigate the contribution of Crp to Y. pestis 231 strain virulence, we infected outbred mice and guinea pigs s.c. with Y. pestis 231, 231Δcrp, or 231P_BAD_-crp. Y. pestis 231P_BAD_-crp strain was grown with arabinose prior to inoculation. The strain 231Δcrp was nonlethal at doses of ≤10^7^ CFU in mice. The LD_50_ of 231Δcrp strain in guinea pigs increased at least by 10^7^-fold (LD_50_ = 2.1 × 10^8^ CFU) compared with that of the wild-type strain (LD_50_ = 3 CFU) (Table 3). The LD_50_s of the 231P_BAD_-crp mutant in mice and guinea pigs were approximately 10^4^-fold and 10^7^-fold higher than those of Y. pestis 231, respectively.

The degree of attenuation of the 231(P_BAD_-crp) mutant strain was increased by eliminating the pPst plasmid. We tested the ability of the plasmid-cured strain to infect outbred mice and guinea pigs. The 231P_BAD_-crp(pPst¯) strain was nonlethal at doses of approximately ≤10^7^ CFU in mice and ≤10^9^ CFU in guinea pigs (Figure 4).

3.3. Humoral Immune Responses

Sera were collected from vaccinated mice and guinea pigs 28 days after immunization. The levels of serum IgG titers against two main immunodominant Y. pestis protective antigens (F1 and LcrV) and Y. pestis whole-cell lysate (YPL) were determined via ELISA (Figure 5). Anti-F1 (p < 0.0001), anti-LcrV (p < 0.0001), and anti-YPL (p < 0.0001) IgG titers in mice and guinea pigs increased in a dose-dependent manner when the levels of IgG induced by different immunized doses were compared. The differences in the levels of IgG to Y. pestis antigens in the sera of mice and guinea pigs immunized with 10^2^ CFU were insignificant (p > 0.05).

3.4. Cell-Mediated Immune Response

Then we measured the levels of the cytokines TNF-α and IFN-γ in sera of mice and guinea pigs at various times after immunization with Y. pestis strain 231P_BAD_-crp(pPst¯) (Figure 6). In s.c.-immunized mice and guinea pigs, induction of pro-inflammatory cytokines IFN-γ and TNF-α followed similar kinetics. The levels of cytokines increased significantly by 48 h. Then the level of IFN-γ in the serum of mice and guinea pigs and TNF-α in guinea pigs decreased. In the serum of mice, the level of TNF-α remained almost at the same level from the second to the sixth day after the immunization with Y. pestis strain 231P_BAD_-crp(pPst¯).

3.5. Ability of s.c. Administered Strain with araC PBAD-Regulated crp to Induce Protective Immunity to s.c. Challenge with Wild-Type Y. pestis Strain 231

Groups of mice and guinea pigs were subcutaneously immunized with 10-fold or 100-fold (respectively) dilution doses of Y. pestis strain 231P_BAD_-crp(pPst¯) and challenged subcutaneously 30 days later with 200 LD_100_ (400 CFU for mice, 6 × 10^3^ CFU for guinea pigs) of the wild-type strain 231.

Mice immunized with 10 CFU of the Y. pestis 231P_BAD_-crp(pPst¯) mutant succumbed to plague after being challenged with 200 LD_100_ of the WT Y. pestis strain 231, but the post-infection lifespan of animals in this group was significantly increased compared with that of the naïve mice. 66.6% of the mice immunized subcutaneously with 10^2^ or 10^3^ CFU survived after challenge with 200 LD_100_ of the WT Y. pestis. 10^2^ CFU, the lowest immunizing dose of the strain 231P_BAD_-crp(pPst¯), did not protect guinea pigs against infection with 200 LD_100_ of the wild type Y. pestis strain 231. The immunizing doses of ≥10^4^ CFU were sufficient to ensure 100% survival of both animal species. (Figure 7).

4. Discussion

Developing live bacterial vaccines presents a significant challenge: ensuring safety without compromising the ability to protect vaccinated individuals [34]. However, traditional attenuation methods often result in bacteria becoming overly susceptible to the host’s innate immunity, which prevents the development of a strong immune response. To overcome these limitations, researchers have engineered bacterial strains that initially exhibit the behavior of virulent wild-type bacteria, allowing them to effectively colonize the host’s tissues. Subsequently, these strains undergo a precisely controlled and delayed attenuation process within the host. This strategy prevents the disease while still triggering a powerful immune response. This regulated delayed attenuation can be accomplished through various genetic modifications [14].

One approach involves substituting promoters of essential for virulence genes (fur, crp, phoPQ, rpoS) with an arabinose-dependent araC P_BAD_ cassette [13]. This ensures that expression of these genes relies on the presence of available arabinose during growth. Upon colonization of host tissues, arabinose deprivation halts the synthesis of these genes’ products, leading to gradual attenuation and preventing disease manifestations.

Initially, the araC P_BAD_ mutant was generated by the developers of the delayed attenuation method. Its LD_50_ for mice was 10^5^ CFU [16], which indicated significant residual virulence and was not good for a vaccine strain. To further reduce virulence, the lpxL gene from E. coli was introduced into the genome of the araC P_BAD_ mutant. This gene is coding a late acyltransferase responsible for the addition of the sixth fatty acid residue to the lipid A. The resulting hexaacyl variant of lipid A is successfully recognized by the host immune system, which prevents the development of a lethal infection [35]. The double mutant further reduced its virulence. For mice its LD_50_ was >10^7^ CFU [36]. Deletion of the introduced gene is not so unlikely. From a biosafety point of view, it is undesirable to use such strains as vaccines, since it is impossible to exclude reverse or compensatory mutations fraught with reversion of virulence. However, a drawback of this vaccine strain variant is the potential loss of the lpxL gene integrated into the Y. pestis genome and restoration of virulence.

To achieve the same goals—further reducing virulence—we used an alternative approach: elimination of the pPst plasmid from the genome, believing that the probability of de novo emergence of the plasminogen activator gene is incomparably lower than the probability of loss of the lpxL gene. This plasmid encodes the well-known pathogenicity factor, plasminogen activator Pla. In the majority of studies, knockout of this gene resulted in an increase in LD_50_ values up to 10^6^ or more CFU [37], but in some experiments LD_50_ values did not increase, while the survival time of mice after infection with a Pla-negative strains slightly increased [38].

Relative evaluation of the Y. pestis Δcrp knockout mutant with its isogenic araC P_BAD_ crp variant with arabinose-dependent regulated expression of the crp gene showed that, compared with the parental KIM5+ strain (bv. Medievalis), both strains were significantly attenuated upon s.c. infection of Swiss Webster mice [16]. LD_50_ of the Δcrp (3 × 10^7^ CFU) and araC P_BAD_ crp (4.3 × 10^5^ CFU) mutants were approximately 10^6^ and 10^4^ times higher than those of the parent Y. pestis strain KIM5+ (<10 CFU), respectively. In our experiments conducted on derivatives of the Y. pestis strain 231 from another phylogenetic group (bv. Antiqua), we obtained similar results. Compared with the parent strain (LD_50_ = 3 CFU for both animals), the LD_50_ of the Δcrp knockout mutant increased for outbred mice and guinea pigs by approximately six and seven orders of magnitude, while the virulence of the delayed-attenuation 231P_BAD_-crp variant decreased by approximately 1.9 × 10^4^ and 1.5 × 10^7^ times.

However, according to the Russian national guidelines [39], Y. pestis candidate vaccine strains should not cause death in mice infected subcutaneously with 10^7^ CFU and guinea pigs infected subcutaneously with 2 × 10^9^ CFU and even 1.5 × 10^10^ CFU. The Δcrp mutant and the less attenuated isogenic P_BAD_-crp mutant did not meet these criteria.

At the same time, it is known that means for achieving regulated delayed attenuation can be combined with other mutations, which together may yield safe efficacious recombinant attenuated bacterial vaccines [36]. The combination of different mutagenesis methods allows uniting diverse strategies for the rational approach to design of attenuated strains of pathogenic bacteria, candidates for vaccine strains [40]. In our study, we took advantage of the fact that in some cases, the loss of the ability to produce plasminogen activator is accompanied by significant attenuation, and the plasmid encoding Pla does not contain genes for protective antigens [38]. After low-temperature cultivation, we were able to select a pPst¯ clone of P_BAD_-crp mutant that reduced its virulence compared to the wild-type strain by at least 7 orders of magnitude in mice and at least 9 orders of magnitude in guinea pigs.

Is the path we have chosen to develop candidate vaccine strains not too complex? Perhaps we could have stopped at one of the intermediate stages? Or, on the contrary, should we have continued editing the genome to remove or modify genes encoding proteins that are undesirable from a vaccinological perspective? Do our mutants combine safety with effective protection for immunized individuals belonging to different species or phylogenetic groups? Below we will try to answer some of these questions.

It is known that different animals react differently to Y. pestis antigens [41]. A mixed vaccine including the strains with special efficacy in rats and in guinea pigs protects both animal species better than the monocomponent ones. A multivalent vaccine, containing strains proven effective against both rats and guinea pigs, offers superior protection for these two animal species compared to vaccines targeting only a single strain. The efficacy of plague vaccines has been demonstrably enhanced through various combinatorial approaches. Studies have shown that administering live attenuated vaccines (#46-S, M #74, or MP-40) in conjunction with the existing EV vaccine significantly amplifies their protective effect [42].

The generation of antibodies specific to Y. pestis is crucial in fighting off infection [39,43,44,45,46,47]. However, a strong and effective immune response against this pathogen also requires the activation of cellular immune mechanisms [48,49,50,51]. Immunization of mice and guinea pigs with 231P_BAD_-crp(pPst¯) elicits anti-F1, anti-LcrV, and anti-YPL serum IgG in a dose-dependent manner. The induction of cellular immunity against Y. pestis by 231P_BAD_-crp(pPst¯) is confirmed by the observed increase in TNF-α and IFN-γ production levels in mice and guinea pigs immunized subcutaneously. Thus, Y. pestis 231P_BAD_-crp(pPst¯) seems to stimulate both humoral and cellular immunity. Mice and guinea pigs immunized subcutaneously with a single dose (10^4^ CFU) of the 231P_BAD_-crp(pPst¯) mutant were completely protected against a subcutaneous challenge with 200 LD_100_ of the wild type Y. pestis strain.

The strain 231P_BAD_-crp(pPst¯) is of interest as a vaccine candidate. It is possible that its predecessor with the pPst plasmid has a greater ability to protect animals due to greater residual virulence, but its ability to cause death in some infected individuals at doses of no more than 10^7^ CFU excludes it from potential vaccine candidates.

5. Conclusions

Our results confirm that arabinose-dependent regulated expression of crp in combination with elimination of the pPst plasmid is an effective strategy for attenuating Y. pestis while maintaining high immunogenicity, providing protection in murine and guinea pig models against bubonic plague.